Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    "-----"

    np_max-act · gemini-2.0-flash

    The neuron activates on email‐style quoting or forwarding markers (lines of dashes or other separators indicating quoted/forwarded message boundaries).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     NRF
    -0.07
     Clarence
    -0.07
    8
    -0.07
     �
    -0.07
    ersistence
    -0.07
     release
    -0.07
     duration
    -0.06
     fear
    -0.06
     Charset
    -0.06
     referral
    -0.06
    POSITIVE LOGITS
    -----
    0.08
     sessiz
    0.07
    键
    0.07
    ecycle
    0.07
    _idx
    0.07
     Barnes
    0.07
     Coleman
    0.07
    ै।↵
    0.07
    ίνει
    0.06
    ipa
    0.06
    Act Density 0.003%

    No Known Activations